Research On The Integration Technology Of Blockchain And Multi-Access Edge Computing Network

Multi-access edge computing(MEC)can realize mobile access,location awareness and low-latency response of large-scale terminal devices,and is a key technology for the development of the next generation Internet.However,there are many challenges in data security,privacy protection,access control,computing offloading,and resource allocation.As a decentralized,immutable,and traceable distributed ledger technology,blockchain provides a novel and effective solution to overcome the above challenges of MEC.The integration of blockchain and multi-access edge computing networks has many advantages,such as enhancing the security of data in MEC,sharing private data between nodes,cross-domain management and control of MEC nodes,decentralized computing offloading,and automatic resource allocation.Therefore,research on the integration technology of blockchain and multi-access edge computing networks is of great significance to the development of safe,reliable and efficient network infrastructure.However,due to the characteristics of decentralization,resource constraints,network heterogeneity,and scale expansion of the MEC network,the integration of blockchain and multi-access edge computing networks has many technical difficulties in the consensus layer,data layer,and application layer.At the consensus layer,existing blockchain consensus algorithms perform poorly in throughput,scalability and energy consumption,making it difficult to meet the requirements of the MEC network.At the data layer,the traditional blockchain ledger structure is bloated and the storage is fully redundant,causing MEC nodes to bear huge storage overhead.At the application layer,the blockchain lacks effective means of accessing and controlling data on the chain,which can easily cause data privacy leakage.In response to the above technical difficulties,this paper studies the integration technology of blockchain and multiaccess edge computing networks from three aspects:consensus algorithm,ledger structure and storage model,and private data sharing.The main work and contributions are as follows:(1)A hybrid consensus algorithm with high scalability and low energy consumption for MECIn order to solve the problems of low throughput,insufficient scalability and high energy consumption of the blockchain consensus algorithm,this paper proposes a highly scalable and low energy consumption hybrid consensus algorithm for MEC based on the characteristics of the MEC network.Firstly,this paper proposes a permissionless consensus algorithm based on proof of capacity to make blockchains permissionless and low energy consumption.MEC nodes compete for block submission rights through capacity space to achieve a green and energy-saving consensus.Secondly,this paper proposes an asynchronous Byzantine-based permissioned consensus algorithm to achieve high throughput of the blockchain.Committee nodes achieve fast consensus on transaction sets through asynchronous reliable broadcast and threshold encrypted voting.Then,this paper mixes the above two consensus algorithms through the structure of "dry leaf chain" and proposes a highly scalable and low-energy hybrid consensus algorithm for MEC to achieve high scalability of the blockchain.The "stem chain"provides security for the election of the "leaf chain" committee and improves the overall scalability of the network.Finally,this paper evaluates the consensus algorithm from the four dimensions of efficiency,decentralization,security and activity through experimental verification.The experimental results show that the throughput of this consensus algorithm can reach 17859.8TPS,the energy consumption is reduced by 64%,and it can support the MEC network with a scale of 2000 nodes.This algorithm meets the requirements of the consensus layer of the MEC network.(2)Lightweight blockchain ledger structure and storage model for MECIn order to solve the problem of excessive blockchain ledger storage overhead,this paper proposes a lightweight blockchain ledger structure and storage model for MEC,which reduces the overall storage overhead of MEC nodes by optimizing the ledger structure and multi-node collaborative storage.First,this paper proposes a state transition proof model based on zero-knowledge proof,which reduces the dependence of block verification on transaction data and enables MEC nodes to independently verify blocks that do not contain transaction data.This greatly reduces block size and improves block verification efficiency.Secondly,this article proposes a multi-node collaborative storage model so that the reliable storage of transaction data is evenly shared by the entire network nodes,reducing the storage overhead of a single point.Then,by integrating the above two models,this paper proposes a lightweight blockchain ledger solution for MEC,which achieves the public verifiability,storage reliability and lightweight single-point storage of the ledger.Finally,this paper develops and deploys a prototype system,and tests this solution from three aspects:storage overhead,block verification efficiency,and block synchronization efficiency.The experimental results show that compared with Ethereum,this scheme can reduce the storage overhead by 53.47%in a single-node network,and can reduce the storage overhead by an order of magnitude in a multi-node network.And as the block height increases,its verification efficiency becomes higher,and the advantages of block synchronization efficiency become more obvious.(3)Blockchain privacy data sharing protocol for MECIn order to solve the problem of privacy protection of data on the chain,this paper proposes a blockchain privacy data sharing protocol for MEC,which achieves decentralized fine-grained encryption and decryption operations by combining the blockchain with the attribute encryption mechanism.Firstly,this paper proposes a decentralized key generation and management protocol to solve the problem of decentralized management of keys in the attribute encryption mechanism.Secondly,this paper proposes a blockchain privacy data sharing protocol for MEC,which enables MEC nodes to flexibly create private data sharing domains and encrypt and decrypt shared data in fine granularity.Then,this paper develops a simulation model of the protocol,tests and analyzes the protocol from three aspects of computational efficiency,scalability and contract execution efficiency.The experimental results show that the encryption and decryption time overhead of this protocol is 787 milliseconds and 128.92 milliseconds respectively,supports a MEC network with a scale of 1599 nodes,and the total contract Gas consumption is only 1313540.Finally,the paper explores the application of this protocol in IoT device identification scenarios.MEC nodes realize the identification of large-scale IoT devices by sharing device fingerprint data.